envimate

AI systems that run in production. And the cloud foundation that keeps them there.

envimate GmbH builds autonomous AI systems and the cloud platforms underneath them. Enterprise cloud since 2016, AI in front of customers since 2019 — on AWS and Azure, with the satellite topics included: cost, security, governance, compliance, and who on your team owns it afterwards.

Book a call nisabek@envimate.com

The method: understand before you automate

Enterprise engagements since 2016 — powerdata · Siemens · mercateo · Gulp · Data Reply · insparx

Offer

What we build

The lead offer

AI across the software lifecycle

We help you put AI to work in every part of software development — writing code, reviewing it, testing it, operating it. And we build software factories: systems of AI agents that take a goal, plan the work, implement, test, and correct themselves — with every cycle evaluated, and people deciding what ships.

That takes more than a model and a prompt: retrieval over your own knowledge, memory that persists between runs, guardrails on what agents may touch, and secure connections between agents and your tools. We build that part too — it is what separates a factory from a demo.

We speak about this work in public: “Beyond Chat: AI Workflows That Actually Investigate Alerts” at WeAreDevelopers World Congress.

The foundation

Cloud platforms & governance

Multi-account AWS and Azure organizations built with infrastructure as code: Terraform, CDK, CloudFormation. Teams create their own projects; the platform keeps them safe.

Budgets with owners. Security as a posture, not a product: guardrails on every account, encryption and access baselines, and findings collected in one place and actually worked off — a state an auditor can verify, not a tool someone installed.

The review

Order in existing systems

“Something is wrong, and nobody can tell us why.” We review systematically — architecture, data, and also how teams and processes are organized.

You get an answer and the reasoning behind it: which technology, where, why, and what it costs.

The method behind all three

Understand before you automate.

Automation built on order survives. Automation built on chaos becomes more chaos, faster.

Case study

The long engagement

powerdata GmbH, energy-industry SaaS provider · 2020–2025

  1. 2020

    The start: 100+ AWS accounts, no order

    The engagement began with one sentence: we have a mess of AWS accounts. More than a hundred of them, across solutions and teams, with no shared structure, no budgets, no governance.

  2. The platform

    Accounts organized per solution and team. Comprehensive multi-team infrastructure as code in Terraform, CDK, and CloudFormation — teams creating new projects themselves. Security Hub governance across the organization. Budgets with owners and a review rhythm across 30 engineers. GDPR documentation (TOMs, VVT) written and maintained alongside. Every SaaS solution — business intelligence, payments, order processing, omnichannel automation — passed AWS architecture reviews.

  3. The product on top

    On that foundation: an omnichannel customer-support automation product for the biggest power and utilities companies in Germany. Phone, email, and chat. PDF documents processed with OCR. Intent recognition with NLP and NLU, including a sophisticated German-language email intent recognition system. Multi-cloud, on AWS and Azure. At scale it resolved around 50,000 calls a day end to end — meter readings, bank changes, payment adjustments — with no human agent involved.

  4. 2024

    The audit that counts

    In 2024 the company was acquired. The platform, the product, and the paperwork went through full due diligence — software analysis, process analysis, organisation analysis. Passed.

The automation was only possible because of the order that was built first. That is the method.

50,000 calls/day 100+ AWS accounts AWS + Azure due diligence passed

Proof

More of the track record

A second cloud, done properly

One of Germany’s largest power and utilities companies ran on Azure and needed AWS established alongside it. We organized the federation, designed an account structure for the scale that came later, and put governance, budgeting, security, and team organisation in from day one.

At home in governed industries

Ten years in power and utilities taught us how governed industries work: data security as an entry requirement, systems that must survive audits, suppliers expected to know the rules before they arrive. Energy, health, mobility, finance — if your industry answers to a regulator, we have built under those constraints before.

Enterprise engagements since 2016

  • Siemens IoT gateway architecture and code review; an istio service-mesh evaluation across AWS and Google Cloud.
  • mercateo A large Spring Boot landscape migrated to AWS: multi-account CI/CD and infrastructure as code.
  • Gulp A security information and event management system (SIEM) on AWS, with an automated code-review toolchain.
  • Data Reply Cassandra cluster automation on AWS with Terraform.
  • insparx Architecture and implementation of a sensitive-data migration.

Engagement model

How we work together

Build it

You have a business problem and need a running system. We design the architecture, implement it with your team, and handle the satellite topics from day one.

Get it under control

Your cloud grew faster than its governance: accounts, costs, security, compliance, unclear ownership. The newest variant is AI-accelerated delivery that outran the team’s understanding of what actually ships. We figure out what is actually there. Then we fix that.

Lead and teach

Fractional CTO or interim technical leadership. We lead, we add senior engineering capacity where your team is stretched, and the point is that your team runs it without us afterwards.

Contract engagements through envimate GmbH, Berlin. Remote or on-site in Germany. English or German.

Book a call

People

Who is envimate

References

What people say

“Nune is one of the most technically proficient CTOs I’ve ever worked with. As an AWS Solutions Architect, I witnessed firsthand her exceptional expertise in a broad range of areas (e.g., AI/ML, DevOps, Cloud Architectures, and Information Security). … She’s very curious to understand her customers’ needs in detail, allowing her to increase value for the customer.”

— Markus Döpke, Senior Solutions Architect at Amazon Web Services (AWS)

“I had the pleasure of working 4 years with Nune as Founder/CIO of powerdata. Her expertise developing modern, reliable, and scalable SaaS products and AI automation drove significant impact in our collaboration and for powerdata’s customers and their satisfaction. … Her leadership and ability to translate complex tech into clear business value and actionable insights make her an exceptional person.”

— Tobias Maier, Team Lead & Senior Account Manager at AWS

“As a cloud architect with great knowledge and fast perception she implemented a new Elasticsearch cluster (based on Docker) hosted on AWS ECS … with Terraform, Ansible and Docker. Also topics like automatic backups and recovery, alerting and security were dealt with great success.”

— Tobias Ott, Technical Account Manager at Amazon Web Services (AWS), as the client

“Her ability to juggle multiple projects was unlike any I’ve seen before and made a dramatic difference in the productivity level of our team. … She knows how to bring the best out of each person and is an outstanding mentor.”

— Anna Martirosyan, Senior Site Reliability Engineer, worked on Nune’s team